The development of chemopreventive agents could be greatly advanced by the use of surrogate endpoint biomarkers that are mechanism based molecular alterations that can be modulated by short-term treatment with chemopreventive agents in parallel with their efficacy to prevent cancer. DNA hypomethylation is an epigenetic and common early event in human and animal tumors that has been proposed for use as a biomarker. We hypothesize that increased DNA demethylase activity is responsible for DNA hypomethylation and that the mechanism by which chemopreventive agents prevent and reverse DNA hypomethylation is to inhibit DNA demethylase activity. We further hypothesize that MBD2 is the DNA demethylase in question. To test our hypotheses, four specific aims are proposed. Specific Aim 1: demonstrate that nongenotoxic carcinogens/tumor promoters induce DNA hypomethylation by increasing DNA demethylase activity; Specific Aim 2: demonstrate that methionine prevents and reverses DNA hypomethylation by inhibiting DNA demethylase activity; and Specific Aim 3 demonstrate that chemopreventive agents reverse DNA hypomethylation in tumors by decreasing DNA demethylase activity. Thus Specific Aims 2 and 3 will test our hypothesis that the mechanism by which chemopreventive agents prevent and /or reverse DNA hypomethylation is by decreasing DNA demethylase activity. Specific Aim 4 will determine if MBD2 is the responsible DNA demethylase. Our long term objective is to validate the use of modulation of DNA hypomethylation as a surrogate end-point biomarker for chemoprevention by demonstrating a common mechanism by which chemopreventive agents of differing pharmacologic actions can prevent and reverse DNA hypomethylation, i.e. decreased DNA demethylase activity. Once modulation of DNA hypomethylation is validated as a surrogate end-point biomarker it could be used to shorten the duration and decrease the cost of clinical trials. [unreadable] [unreadable] [unreadable]